Data assimilation using recurrent radial basis function neural network model

The capabilities of existing computational modelling technologies are continuously advancing and integration of data and modelling techniques is nowadays receiving enormous attention. Advances in data storage and its retrieval have been enormous in recent years. The water related issues due to urbanization, population and economic growth are becoming more and more complex and require every technique to be employed to its fullest limits in order to achieve sustainable water resources management. As a response to these challenges, a novel data assimilation approach integrating the deterministic model and the neural network model with the measured data is described in this paper and it is proved to be much more powerful than the traditional modelling approach based on utilising the deterministic model alone. This approach has shown to be capable of achieving higher model accuracies and better knowledge about the state of a hydrodynamic system.